Snoop filtering system in a multiprocessor system

A system for supporting cache coherency in a computing environment having multiple processing units, each processing unit having associated cache memory operatively coupled therewith, said system comprising:

• a plurality of interconnected snoop filter units, each snoop filter unit corresponding to and in communication with a respective processing unit, with each snoop filter unit comprising;
• a plurality of receiving means each for receiving asynchronous snoop requests from a respective memory writing source in said multiprocessor computing environment;
• and,
• a plurality of parallel operating filter means coupled in one-to-one correspondence with each receiving means for processing snoop requests received thereat and one of forwarding snoop requests or preventing snoop requests from being forwarded to its associated processing unit, each of said plurality of parallel operating filter means comprising parallel operating sub-filter elements, each simultaneously receiving an identical snoop request and implementing one or more different snoop filter algorithms for determining those snoop requests for data that are determined not cached locally at said associated processing unit and preventing forwarding of those requests to said processor unit,
• whereby a number of snoop requests forwarded to a processing unit is reduced thereby increasing performance of the computing environment.

A snoop filtering method for a computing environment having multiple processing units, each processing unit having one or more cache memories associated therewith, said method comprising:

• providing a snoop filter sub-system associated with each of said multiple processing units, each snoop filter sub-system having a plurality of dedicated inputs for receiving snoop requests from dedicated memory writing sources in said multiprocessor computing environment and,
• interconnecting each said snoop filter sub-system in a point-to-point fashion by providing communication links for directly connecting a memory writing source to a corresponding dedicated input of each snoop filter sub-system associated with each processing unit of said multiprocessor computing environment;
• wherein each said snoop filter sub-system implements steps of:
• providing a plurality of snoop filter means each coupled to a corresponding one of said plurality of dedicated inputs for filtering respective received snoop requests from respective dedicated memory writing sources, each snoop filter means implementing one or more parallel operating sub-filter elements for:
• determining those requests likely residing in a cache memory at said associated processing unit;
• and,
• filtering out those snoop requests for data that are determined not cached locally at said associated processing unit,
• whereby a number of snoop requests forwarded to a processing unit is significantly reduced thereby increasing performance of said computing environment.